Real-time synchronization is one of the essential theories in real-time systems, and the recent booming of parallel real-time tasks has brought new challenges to the synchronization analysis. As the easy implementation and negligible overheads, spin-locks have received much interest since the study for sequential tasks. However, existing spin-based blocking analyses for parallel tasks are relied on execution-time inflation, and the substantially more accurate inflation-free analysis has not been fathomed yet. Moreover, existing analyses suffer an overrepresentation problem, which can be further exacerbated for parallel tasks with spin-locks. To overcome such pessimism, we propose an improved blocking analysis for non-preemptive spin-locks based on a finer-grained shared resource model. In particular, we consider individual length for each shared resource request and use the state-of-the-art linear optimization technique to achieve a pinpoint inflation-free analysis. Empirical evaluations show that the proposed analysis dominated other state-of-the-art analysis, which further shows the improved accuracy achieved by the proposed approach.

Real-time scheduling and locking protocols are fundamental facilities to construct time-critical systems. For parallel real-time tasks, predictable locking protocols are required when concurrent sub-jobs mutually exclusive access to shared resources. This paper for the first time studies the distributed synchronization framework of parallel real-time tasks, where both tasks and global resources are partitioned to designated processors, and requests to each global resource are conducted on the processor on which the resource is partitioned. We extend the Distributed Priority Ceiling Protocol (DPCP) for parallel tasks under federated scheduling, with which we proved that a request can be blocked by at most one lower-priority request. We develop task and resource partitioning heuristics and propose analysis techniques to safely bound the task response times. Numerical evaluation (with heavy tasks on 8-, 16-, and 32-core processors) indicates that the proposed methods improve the schedulability significantly compared to the state-of-the-art locking protocols under federated scheduling.